Serum amyloid P component (SAP) is a normal, structurally invariant, soluble, non-fibrillar, constitutive plasma glycoprotein, mass 127,310 Da, produced exclusively by the liver. It is composed of 5 identical 25,462 Da protomers non-covalently associated with cyclic pentameric symmetry in a disc like configuration. Each subunit, composed of a flattened β-jellyroll, with tightly tethered loops joining the β-strands, contains a single calcium dependent ligand binding site on the planar B (binding) face of the intact pentamer. SAP binds to all types of amyloid fibrils with the high avidity conferred by multivalent interactions. This strictly calcium dependent interaction is responsible for the universal presence of human SAP in all human amyloid deposits of all types, and hence the name of the protein, where P stands for plasma, the source of this component of amyloid. In addition to its capacity for specific calcium dependent binding to particular ligands, a key property of human SAP is that the protein itself is intrinsically resistant to proteolysis. Its avid binding to amyloid fibrils is mutually stabilising, strongly protecting the fibrils against proteolysis and phagocytic degradation in vitro1 and significantly contributing to persistence of amyloid in vivo2. These observations underlie the validation of SAP as a therapeutic target in amyloidosis (M B Pepys & T L Blundell, U.S. Pat. No. 6,126,918, 3 Oct. 2000). Furthermore, binding of SAP to nascent amyloid fibrils strongly promotes amyloid fibrillogenesis3-5. European patent application EP 0915088 discloses compounds which are competitive inhibitors of binding of SAP to amyloid fibrils, as well as methods for their manufacture. One of the compounds disclosed therein is (R)-1-[6-[(R)-2-carboxy-pyrrolidin-1-yl]-6-oxo-hexanoyl]pyrrolidine-2-carboxylic acid (CPHPC).
Administration of these palindromic bivalent ligands for SAP causes the rapid and almost complete depletion of SAP from the circulation for as long as the compounds are administered6,7, as described in WO2003/013508, U.S. Pat. No. 7,045,499; U.S. Pat. No. 7,691,897; and U.S. Pat. No. 8,173,694. This treatment also reduces the amount of SAP associated with the amyloid deposits but does not remove all the SAP′.
Amyloid is an abnormal, insoluble, extracellular deposit, composed predominantly of characteristic protein fibrils8 together with abundant proteoglycans and glycosaminoglycans. About 25 different, unrelated, natively soluble, globular proteins form the amyloid fibrils which cause the different types of systemic amyloidosis but all amyloid fibrils have very similar morphology and the same cross-β core structure. This structure binds the ordered arrays of Congo red dye molecules which create pathognomonic red-green birefringence in strong cross polarised light: the gold standard diagnostic criterion for amyloid. Certain soluble, non-fibrillar plasma proteins may also be present in amyloid deposits but only one, serum amyloid P component (SAP), is universal in all human amyloid deposits, due to its avid, specific, calcium dependent binding to all types of amyloid fibrils.
Amyloid deposits disrupt the structure and function of affected tissues and organs, causing the serious disease, amyloidosis. Systemic amyloidosis is a rare, fatal condition caused by amyloid deposits which may be present in connective tissue and blood vessel walls throughout the body, as well as the parenchyma of the major organs, but never in brain substance itself. In local amyloidosis, the amyloid deposits are confined to a single anatomical site or a single organ/tissue system. Cerebral amyloid angiopathy, with amyloid deposition confined to the walls of cerebral blood vessels, is the most common and important form of local amyloidosis. It is responsible for a substantial proportion of intracerebral haemorrhages in both demented Alzheimer's disease patients and non-demented individuals, and is thus an important cause of dementia in its own right.
Treatment of systemic amyloidosis patients with CPHPC produced almost complete depletion of circulating SAP for as long as the drug was administered but did not remove all the SAP bound to the amyloid deposits.7 The patients remained clinically stable while being treated, with no new amyloid accumulation, and their organ function was maintained but there was no regression of amyloid, probably due to the failure of complete removal of amyloid bound SAP. Since the amyloid deposits in the tissues are the direct cause of disease, it is highly desirable that they should be eliminated. This important unmet medical need led to the invention of a new approach to treatment of amyloidosis in which SAP bound to amyloid deposits is used as a target for anti-human SAP antibodies. Such antibodies could not be safely or effectively administered to patients with normal circulating concentrations of SAP since the antibodies would bind to the SAP in the plasma, forming tissue damaging immune complexes, and the antibodies would be consumed in this process making them unavailable for binding to SAP in amyloid. However prior administration of CPHPC depletes SAP from the circulation, so that anti-SAP antibodies can be safely infused and remain available to bind to residual SAP left in the amyloid deposits. Binding of the antibodies to the amyloid-associated SAP activates the complement system and engages macrophages to phagocytose and destroy the amyloid deposits leading to clinical benefit.
International patent application WO2009/000926 discloses the use of compounds which deplete SAP from the circulation co-administered with an antibody specific for SAP for potential treatment of amyloidosis.
International patent application WO2009/155962 discloses mouse monoclonal antibody Abp1 and provides binding and efficacy data for various mouse monoclonal antibodies which may be co-administered with compounds which deplete SAP from the circulation for potential use in the treatment of amyloidosis.
International patent application WO2011/107480 discloses antigen binding proteins, in particular humanised antibodies, specific for SAP and their potential use in the treatment of diseases associated with amyloid deposition.
In addition to the rare clinical condition of amyloidosis, which is unequivocally directly caused by extracellular amyloid deposition disrupting tissue structure and function, amyloid deposits are also present in two very common and important diseases: Alzheimer's disease and type 2 diabetes. In these latter diseases, the amyloid deposits are microscopic, are confined to the brain and islets of Langerhans respectively, and it is not known whether or how they may contribute to pathogenesis of neurodegeneration and diabetes respectively. Alzheimer's disease and type 2 diabetes thus cannot be classified as forms of amyloidosis but rather should be considered as amyloid associated diseases. Nevertheless, amyloid deposits are always present in them and the deposits also always contain SAP15-19. The brain in Alzheimer's disease also contains another type of abnormal insoluble fibrillar protein aggregate, known as neurofibrillary tangles, and SAP binds avidly to these, as it does to amyloid.15-19 Neurofibrillary tangles bearing SAP, but not amyloid, are present in the brain in other types of dementia, including the frontotemporal dementias
In addition to, and quite independent of, its role in amyloidosis, human SAP binds to and enters cerebral neurones and causes neuronal apoptosis in vitro and in vivo9-13. It has been shown that unique, pharmaceutical grade pure human SAP14 disrupts synaptic transmission, causing abnormal paired pulse ratio and long term potentiation in organotypic rodent brain slices in vitro.
The cerebral neurotoxicity of human SAP is therefore likely to contribute to neurodegeneration in humane9-12, 20. The fact that most of the common risk factors for dementia increase brain exposure to SAP is consistent with this concept. Thus age, a key risk factor, is associated with prolonged exposure of the ageing brain to normal SAP concentrations, whilst the major risk factors of non-penetrating head trauma and cerebral haemorrhage cause blood to enter the brain, sharply increasing cerebral SAP content. In Alzheimer's disease, brain content of SAP is abnormally high due to its binding to amyloid deposits and neurofibrillary tangles15-19. This is also likely to be the case in other dementias with neurofibrillary tangles but not amyloid deposits. Importantly, higher brain SAP content is reported in demented Alzheimer's disease patients than in elderly subjects who were cognitively intact at death, either with or without co-existing plaques and tangles at autopsy20. The significant positive correlation between cerebral SAP content and dementia20 is consistent with a causative role for SAP.
The quantities of SAP in cerebrospinal fluid and bound to cerebral amyloid deposits and neurofibrillary tangles are dramatically lower than in systemic extracellular fluid and on systemic amyloid deposits respectively. Depletion of plasma SAP by CPHPC, from the normal 20-50 mg/L to <0.1 mg/L, reduces the CSF SAP concentration from 2-30 μg/L to <0.1 μg/L in patients with Alzheimer's disease21. Human SAP is produced only by the liver and reaches the brain via the blood22. CPHPC treatment thus removes virtually all SAP from the cerebrospinal fluid and, since SAP binding is fully reversible, will also remove it from the cerebral amyloid deposits and neurofibrillary tangles. Furthermore in Alzheimer's disease patients, CPHPC enters the cerebrospinal fluid21 where it can also block binding of any free SAP to amyloid fibrils, to neurofibrillary tangles and cerebral neurones. All amyloid fibril types can be degraded by proteases and phagocytic cells in vitro1, and systemic amyloid deposits spontaneously regress slowly in vivo when the abundance of their respective fibril precursor proteins is sufficiently reduced8. Mechanisms for amyloid clearance thus do operate in vivo. Confirmation that human SAP is itself neurocytotoxic9-13, independent of its binding to amyloid, demonstrates the potential, additional, direct benefit of SAP depletion.
All plasma proteins enter diseased or damaged joints in vivo but in patients with various different arthropathies, uptake of radiolabelled SAP into some joints that did not have clinically detectable effusions was observed. Furthermore the concentration of SAP in synovial effusion fluids was substantially below that predicted from the molecular size of SAP, demonstrating that the SAP visualised scintigraphically was not free in solution but was actually bound to solid structures within the joint. Synovium, articular cartilage and/or joint capsules of elderly individuals often contain microscopic amyloid deposits, associated with age rather than extent or severity of osteoarthritis, and these deposits could explain the localisation of SAP. SAP also binds avidly, in vivo as well as in vitro, to exposed DNA and chromatin, and to apoptotic cells. Increased cell death in osteoarthritic joints may thus provide ligands for SAP deposition. WO2004/108131 discloses the treatment of patients with osteoarthritis by injection of CPHPC ((R)-1-[6-[(R)-2-carboxy-pyrrolidin-1-yl]-6-oxo-hexanoyl]pyrrolidine-2-carboxylic acid) resulting in the alleviation of osteoarthritis symptoms.
Human SAP binds avidly to all forms of free DNA and also to chromatin, both in vitro and in vivo. Indeed SAP is the only normal human plasma protein which binds specifically in a calcium dependent interaction with DNA23,24. In contrast, SAP from some other species, including mouse and horse, binds weakly if at all to DNA, and dogs and rabbits do not even have the SAP gene and thus produce no SAP. DNA vaccination, in which immunisation is achieved by injection of DNA coding for the immunogen rather by injection of the immunogen itself, has been extensively investigated as a highly desirable approach to induction of protective immunity against infections and as a potential immunotherapeutic intervention in cancer. However, although DNA vaccination is effective in mice. dogs, rabbits and horses, it has consistently failed in humans and also in cows, which like humans have SAP which binds avidly to DNA. In the species in which DNA vaccination works, SAP either does not bind to DNA or is absent. Furthermore, experiments in mice with transgenic expression of human SAP and using CPHPC to deplete it, confirm that the presence of human SAP blocks efficacy of DNA vaccination.25,26 
SAP binds to some bacterial species but not to others. For those pathogenic bacteria to which SAP binds, the SAP has a powerful anti-opsonic effect in vitro and in vivo, reducing phagocytosis and killing of the organisms and thus protecting them from the host's innate immune system27. This effect, which promotes infectivity and virulence, is abrogated by administration of CPHPC to inhibit binding of SAP to the bacteria27. SAP is also present bound to the surface of the fungal cells in the tissues of patients suffering from invasive candidiasis28. This binding reflects the presence of amyloid fibrils, formed from fungal proteins, on the surface of the pathogenic organism.28 
CPHPC is pharmacologically effective but it has very low and variable oral bioavailability of ˜3-5% and therefore only parenteral administration by intravenous infusion or subcutaneous injection is optimal to achieve the desired SAP depletion. However most of the existing and potential indications for therapeutic SAP depletion require long term administration. Long term intravenous administration is not practical. Although long term subcutaneous administration is feasible, the injections may cause stinging discomfort and this is not tolerated by some patients.7 
WO2003/051836 discloses D-prolines pro-drugs useful for the treatment of diseases where SAP depletion has a beneficial effect. The 25 Examples disclosed therein were predominantly obtained as oils; only 5 of them were solid. In proceedings at the European Patent Office, a divisional application to the European equivalent of WO2003/051836 was filed with claims directed to (R)-1-(6-{(R)-2-[1-(2,2-dimethyl-propionyloxy)-ethoxycarbonyl]-pyrrolidin-1-yl}-6-oxo-hexanoyl)-pyrrolidine-2-carboxylic acid 1-(2,2-dimethyl-propionyloxy)-ethyl ester. At the time of filing. the GlaxoSmithKline group of companies (Glaxo Group Limited) has a Licence and Research Collaboration Agreement with Pentraxin Therapeutics Limited including a licence to EP 0915088 and WO2003/051836, and corresponding applications thereof.
There is therefore a need for a compound which is capable of generating CPHPC in quantities capable of depleting SAP efficiently following oral administration, whilst possessing physicochemical properties suitable for pharmaceutical development.